Practical Deep Learning at Scale with MLflow: Bridge the gap between offline experimentation and online production

Yong Liu

€8.99 ~~€28.99~~

4.5 (11 Ratings)

eBook Jul 2022 288 pages 1st Edition

Yong Liu

€8.99 ~~€28.99~~

4.5 (11 Ratings)

eBook Jul 2022 288 pages 1st Edition

What do you get with eBook?

Instant access to your Digital eBook purchase

Download this book in EPUB and PDF formats

Access this title in our online reader with advanced features

DRM FREE - Read whenever, wherever and however you want

AI Assistant (beta) to help accelerate your learning

View table of contents

Preview Book

Download Code

Key benefits

Focus on deep learning models and MLflow to develop practical business AI solutions at scale
Ship deep learning pipelines from experimentation to production with provenance tracking
Learn to train, run, tune and deploy deep learning pipelines with explainability and reproducibility

Description

The book starts with an overview of the deep learning (DL) life cycle and the emerging Machine Learning Ops (MLOps) field, providing a clear picture of the four pillars of deep learning: data, model, code, and explainability and the role of MLflow in these areas. From there onward, it guides you step by step in understanding the concept of MLflow experiments and usage patterns, using MLflow as a unified framework to track DL data, code and pipelines, models, parameters, and metrics at scale. You’ll also tackle running DL pipelines in a distributed execution environment with reproducibility and provenance tracking, and tuning DL models through hyperparameter optimization (HPO) with Ray Tune, Optuna, and HyperBand. As you progress, you’ll learn how to build a multi-step DL inference pipeline with preprocessing and postprocessing steps, deploy a DL inference pipeline for production using Ray Serve and AWS SageMaker, and finally create a DL explanation as a service (EaaS) using the popular Shapley Additive Explanations (SHAP) toolbox. By the end of this book, you’ll have built the foundation and gained the hands-on experience you need to develop a DL pipeline solution from initial offline experimentation to final deployment and production, all within a reproducible and open source framework.

Who is this book for?

This book is for machine learning practitioners including data scientists, data engineers, ML engineers, and scientists who want to build scalable full life cycle deep learning pipelines with reproducibility and provenance tracking using MLflow. A basic understanding of data science and machine learning is necessary to grasp the concepts presented in this book.

What you will learn

Understand MLOps and deep learning life cycle development
Track deep learning models, code, data, parameters, and metrics
Build, deploy, and run deep learning model pipelines anywhere
Run hyperparameter optimization at scale to tune deep learning models
Build production-grade multi-step deep learning inference pipelines
Implement scalable deep learning explainability as a service
Deploy deep learning batch and streaming inference services
Ship practical NLP solutions from experimentation to production

What do you get with eBook?

Instant access to your Digital eBook purchase

Download this book in EPUB and PDF formats

Access this title in our online reader with advanced features

DRM FREE - Read whenever, wherever and however you want

AI Assistant (beta) to help accelerate your learning

Frequently bought together

€35.99

€38.99

Machine Learning Engineering with MLflow

€32.99

Total € 107.97

Filter reviews by

All

Packt verified reviews

Amazon verified reviews

MYLiang Jul 08, 2022

This book gives a good introduction to the full life cycle of deep learning development using MLflow. It's easy to read, and also very practical with all the code provided. The book illustrates the challenges in every step of developing and productionizing a deep learning model, and how you can deal with these challenges with MLflow, at scale and with better explainability. Recommended for everyone who works with data and want a better management of your models or projects with MLflow. You'll benefit from and be inspired by different sections of this book.

Amazon Verified review

Andrew J. Brooks Jul 25, 2022

The beauty of this work is that it threads together the most important concepts facing ML practitioners today AND actionable recommendations for tooling from the modern tech stack AND working code. There are tutorials and blog posts out there that touch on bits and pieces of these, but they are fragmented with many holes in between. The in-depth guide provided within this work connects the intuition and full ML lifecycle to these concepts in a way that the field desperately needs.Having worked directly with Yong for several years on many of these topics, I can attest that this book is not just a tome of facts and tutorials, but a trove of wisdom developed through years of experience and experimentation. For example, even seasoned ML Practitioners will likely find new insights and patterns in Chapter 7 for how to elegantly connect model and business and pre/post-processing logic that is often disjointed. Or the landscape of options and considerations for serving MLFlow models in Chapter 8. The techniques covered in this text are highly practical for anyone shipping industry-grade ML, but rarely covered with this much depth (or at all) elsewhere on the web.

Joe Jul 10, 2022

Anyone jumping onto the ML wagon of any kind will reap the benefits of a massive head-start from this book. The book is written by a data scientist for data scientists and ML engineers. It is a cookbook filled with small, medium, and large code snippets and practical insights covering the whole spectrum of ML dev and ops. The tone of the book is very conversational yet instructive at the same time.I would recommend this book to any Machine Learning team with whom I have ever worked. Agile teams have the best chance to get the most value from Yong because they could start by following a best-tried-out path. Still, the established engineering unit can utilize many best practices presented in the book instead of spending thousands of hours trying to decipher cryptic documentation from vendors. (We all know how it feels when navigating through tons of websites and trying to find a needle in a haystack.)

QAM Chen Jul 09, 2022

Databricks with mlflow server is a very well setup environment for MLOps. The book gives you all knowledge you need for model building and MLOps. You can use it for traditional ML training or NLP transfer learning model building. I learned a lot from the book especially model deployment and model hyper-params tune.Overall: Reading the book and you can learn ML lifecycle with databricks and MLflow which is very important for model to production.

Young C. Jul 10, 2022

This book gave a great coverage of MLflow and deep learning centric toolsets with practiced examples. The author obviously took great efforts on both writing and making sure the provided examples are up-to-date and working, by combing his industry experience and insights.MLflow is one of the major tool for managing ML modelling, inference and monitoring lifecycle with relevant artifacts which can boost the productivity of your work for making your model , experiments and systems managed and reproducible. This book could be used as a good guidance for entering real world ML and equip yourself with both some foundation of ML lifecycle and the tool usages.I could reserve a star from the rating, but it's not necessarily about this specific book. Just want to emphasize that in software world, things are changing fast and there are often more than one choices to do the same thing. Due to limitation of the time and length of the book, you can't rely on a single book to be professional enough to deal with real world challenges. Keep learning and practicing are still needed.

Practical Deep Learning at Scale with MLflow: Bridge the gap between offline experimentation and online production

What do you get with eBook?